INDEX

Explanations

words related to a decrease or reduction in something

oai_token-act-pair · gpt-3.5-turbo

words indicating depletion or reduction

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

sw

-0.69

 covers

-0.64

SW

-0.59

 Astron

-0.59

ples

-0.58

 lights

-0.57

serv

-0.56

aly

-0.56

 Newman

-0.55

nex

-0.55

POSITIVE LOGITS

hift

1.01

ometimes

0.99

hirt

0.98

heet

0.93

imentary

0.90

omething

0.88

peak

0.87

anamo

0.84

pread

0.84

terday

0.82

Activations Density 0.138%

words related to a decrease or reduction in something

words indicating depletion or reduction

No Comments

No Known Activations